January 1, 2014 (Vol. 34, No. 1)
Methodological Approach Designed for Ease of Use, Speed, and Low Cost to Clone
CRISPR endonucleases have been used in a recent flood of literature due to their elegant and simple mode of RNA-guided gene targeting and ability to operate via protocols developed originally for ZFN-based genome editing in human cells, rats, flies, worms, and plants among other organisms.1-4
The literature also suggests the simplicity and speed of the current CRISPR targeting approach has come at the price of specificity. Minimization of off-target effects can be particularly challenging for highly repetitive regions of mammalian genomes and requires forethought on the best choice of genome editing technology.
Are there cases when ZFNs or CRISPRs are more appropriate? Should the nuclease be delivered as plasmid or mRNA? Should the donor molecule be a small single-stranded oligo or a larger double-stranded plasmid? This article serves as an abridged guide to Sigma-Aldrich CRISPR genome editing, design, and workflows for deriving specifically modified cells.
CRISPR Design and Specificity
Several groups demonstrated in 2013 that CRISPR endonucleases, although generally robust, have shown wide variation in activity, even among CRISPRs targeting loci in close proximity.2 Thus, testing three to four CRISPR nucleases targeting different DNA sequences is prudent. As the length of the DNA target site for CRISPR systems (19–20 bp) is significantly smaller than that of CompoZr® ZFNs (e.g., 30–36 bp) or TALENs incorporating engineered nucleases that mandate heterodimerization, care must be taken during design to ensure minimal off-target breaks elsewhere in the genome (Figure 1).
Recent evidence indicates that off-targeting by CRISPR endonucleases is a significant concern and some guidance for avoiding off-target activity is beginning to develop.5,6 For example, there is general agreement that at least 3 bp of mismatch in a single CRISPR target site (vs. the target genome) provides suitable specificity for research applications.
Online tools, such as the Sigma CRISPR bioinformatics online design tool, can apply specificity rules automatically to identify appropriate CRISPR designs for targeting sequences in the human, mouse, and rat genomes. To further tackle CRISPR off-target effects, paired Cas9-nickase approaches are being explored that mimic the heterodimeric architectures of FokI-based ZFNs and TALENs.4,7
Figure 1. As target site length in the human genome increases, the percentage of unique sites increases and the chance of off-target breaks decreases. The length of the DNA target site for single-site CRISPR nucleases is significantly smaller than that of CompoZr ZFNs, meaning that attention must be paid to minimize off-target breaks elsewhere in the genome during the design process.
Delivery of CRISPR Plasmids
For initial experiments in human cells, the expedient route is nucleofecting maxi-prepped CRISPR plasmid DNA into well-validated cell types such as K562 or U2OS to assess double-stranded break (DSB) activity or donor integration levels. These cell lines have been shown to respond with high levels of homologous recombination, which is ideal for testing nuclease compatibility with new vectors before exploring genome editing in a challenging or unknown cell type.
For mouse and rat cell culture testing, neuro2A and C6 are suitable cell types for initial CRISPR experiments. A starting point for dosage experiments is 2–8 µg (total) of CRISPR and guide RNA plasmids per 0.5 to 1 million cells.
When delivery of dsDNA would be acutely toxic, for example in dendritic cells9, transgene delivery as mRNA often proves successful. This has been demonstrated for ZFNs in cell culture and animal applications and, thus far, for CRISPRs in mice. Various cell types mount unique responses to exogenous DNA. Thus both plasmid and mRNA-based delivery of ZFN and CRISPR endonucleases should be explored when using a cell type lacking published protocols for genome editing. Transfection efficiency can be evaluated by microscopy or FACS, as the Sigma CRISPR single vector format contains a GFP (or RFP)-linked expression cassette for Cas9.
Design of Donor DNA
When operating by themselves in mammalian cells, targeted endonucleases make destructive double-strand breaks (DSBs), which often result in deletions of ~1 to 20 bp and, to a lesser extent, insertions. These nuclease-only modifications are an efficient and practical approach for gene knockout. To make delicate single base changes, a donor DNA is required to assist the broken chromosome by serving as a repair template. Donor DNAs can function successfully in many forms by two primary modes: (1) copying new DNA information via homology-based mechanisms and (2) direct ligation to DSB ends.
The homology-based donors typically use plasmid templates that encode the desired change plus ~800 bp of homologous sequence on either side of the break site.10-12 Small single-stranded oligonucleotides have also been shown to function effectively as homology-based donors when the mutation of interest is in close proximity to the cut site.13,14 Both strategies have been applied successfully with CRISPR systems.15
End-ligation donor linear dsDNA fragments and plasmids allow irreversible transgene insertion, treating an entire chromosome as a ligation-ready DNA fragment.16-18 One technique involves transfecting into mammalian cells plasmids harboring an inverted ZFN cut site and ligating these fragments to chromosomal breaks induced by the same ZFN and that feature compatible overhangs. Due to the inversion of the cut sites, the recombined plasmid-chromosome hybrid lacks the nuclease cut site and cannot be re-cleaved. Based on current donor integration mechanistic data, direct ligation is expected to support more efficient insertion of large DNA fragments, including those in excess of 10 kb.
The effectiveness of end-ligation donors with CRISPR systems depends on the type of break created. While first-generation CRISPR systems use a single Cas9 protein to induce a mostly blunt-ended DSB (known to be less efficient for in vitro ligation than sticky-ends), the efficiency of intracellular nuclear blunt ligation of donor DNA to chromosomes remains to be characterized.
Monitoring CRISPR DSB Activity
Several published protocols exist for monitoring DSB activity. The most rapid, flexible, and economical assay is the mismatch cleavage assay, which can detect the indels generated by non-homologous end joining activity in eukaryotic cells. The most common protocols use the CEL-I enzyme (a.k.a. Surveyor nuclease)19 and T7 endonuclease I (T7EI)20. Provided ample resources, deep sequencing can also be used to detect and quantitate indel activity in CRISPR-treated cell populations. If DSB activity cannot be detected despite many attempts for a particularly favored CRISPR design, consider enriching the cell population for Cas9-FP expression as described below.
Single-Cell Cloning and Genotyping
Since Sigma CRISPR single-vector format plasmids contain an FP-linked Cas9 expression cassette, fluorescence-activated cell sorting (FACS) can be used to isolate cell populations with significantly increased frequencies of Cas9-induced modifications. The Cas9 and FP protein coding regions are linked by a small sequence encoding a self-cleaving peptide, which utilizes “ribosomal skipping” to generate two individual proteins.
This FACS-enrichment approach is particularly useful in scenarios in which delivery efficiencies and/or Cas9 expression levels are low or undetectable. Furthermore, Cas9-FP linked expression eliminates the extra step of cloning artificial target sites into surrogate reporter plasmids. It also reduces the total amount of transfected dsDNA, which can often result in cell toxicity. Finally, it opens the option to use “dsDNA free” approaches by implementing Cas9-FP mRNA, in vitro transcribed gRNA, and single stranded oligos.
Cells can be harvested for FACS according to commonly used protocols. Cells can be sorted into fractions with low, medium, and high FP expression levels. Cell populations with the highest FP expression have been shown to enrich genome edits with the extent of the “top” fraction ranging from 1 to 60% of the total cell population (Figure 2).
Figure 2. Methods to monitor and enrich cells harboring active FP-linked Cas9 expression cassettes. (A) Fluorescence microscopy (left panel) or FACS (right panels) can be used to monitor the delivery of Cas9-FPs. (B) Cell fractions divided by level of FP expression show corresponding increases in CRISPR-associated DSB activity at KRAS and CCR5 target sites.
Selecting the Right Nuclease
A large collection of published literature based on thousands of CompoZr ZFN designs is useful for modeling and adapting methods. That said, not all genomic loci are easy to engineer. For example, the HBB locus in human cells has proven difficult to engineer using both ZFNs and TALENs.21
CRISPR systems present the opportunity to determine project feasibility because they are easy and inexpensive to clone and use to screen compatible donor DNA and genomic target sites. If specificity is a concern, experiments can be duplicated with a higher-fidelity nuclease, such as CompoZr ZFNs, that targets the same region.
Greg Davis ([email protected]) is a principal R&D scientist and Christian Nievera ([email protected]) is a product manager at Sigma Life Science.
References
1. Jinek. M., et al., A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 2012; 337(6096): 816-21.
2. Cong, L., et al., Multiplex genome engineering using CRISPR/Cas systems. Science 2013; 339(6121):819-23.
3. Mali, P., et al., RNA-guided human genome engineering via Cas9. Science 2013; 339(6121):823-6.
4. Mali P, Aach J, Stranges PB, Esvelt KM, Moosburner M, Kosuri S, Yang L, Church GM. CAS9 transcriptional activators for target specificity screening and paired nickases for cooperative genome engineering. Nat Biotechnol. 2013 Sep;31(9):833-8. doi: 10.1038/nbt.2675. Epub 2013 Aug 1. PubMed PMID: 23907171; PubMed Central PMCID: PMC3818127.
5. Fu, Y., et al., High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nat Biotechnol. 2013. doi: 10.1038/nbt.2623. PubMed PMID: 23792628.
6. Hsu, P. D., et al., DNA targeting specificity of RNA-guided Cas9 nucleases. Nat Biotechnol. 2013. doi: 10.1038/nbt.2647. PubMed PMID: 23873081.
7. Ran FA, Hsu PD, Lin CY, Gootenberg JS, Konermann S, Trevino AE, Scott DA, Inoue A, Matoba S, Zhang Y, Zhang F. Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity. Cell. 2013 Sep 12;154(6):1380-9. doi: 10.1016/j.cell.2013.08.021. Epub 2013 Aug 29. PubMed PMID: 23992846.
8. Sigma-Aldrich, unpublished data
9. Van Tendeloo VF, Highly efficient gene delivery by mRNA electroporation in human hematopoietic cells: superiority to lipofection and passive pulsing of mRNA and to electroporation of plasmid cDNA for tumor antigen loading of dendritic cells. Blood. 2001 Jul 1;98(1):49-56.
10. Bibikova M, Beumer K, Trautman JK, Carroll D. Enhancing gene targeting withdesigned zinc finger nucleases. Science. 2003 May 2;300(5620):764. PubMed PMID: 12730594.
11. Urnov FD, Miller JC, Lee YL, Beausejour CM, Rock JM, Augustus S, Jamieson AC, Porteus MH, Gregory PD, Holmes MC. Highly efficient endogenous human genecorrection using designed zinc-finger nucleases. Nature. 2005 Jun 2;435(7042):646-51. Epub 2005 Apr 3. PubMed PMID: 15806097.
12. Cui X, Ji D, Fisher DA, Wu Y, Briner DM, Weinstein EJ. Targeted integration in rat and mouse embryos with zinc-finger nucleases. Nat Biotechnol. 2011 Jan;29(1):64-7. doi: 10.1038/nbt.1731. Epub 2010 Dec 12. PubMed PMID: 21151125.
13. Chen F, Pruett-Miller SM, Huang Y, Gjoka M, Duda K, Taunton J, Collingwood TN, Frodin M, Davis GD. High-frequency genome editing using ssDNA oligonucleotides with zinc-finger nucleases. Nat Methods. 2011 Jul 17;8(9):753-5. doi:10.1038/nmeth.1653. PubMed PMID: 21765410; PubMed Central PMCID: PMC3617923.
14. Meyer M, Ortiz O, Hrabé de Angelis M, Wurst W, Kühn R. Modeling disease mutations by gene targeting in one-cell mouse embryos. Proc Natl Acad Sci U S A. 2012 Jun 12;109(24):9354-9. doi: 10.1073/pnas.1121203109. Epub 2012 Jun 1. PubMed PMID: 22660928; PubMed Central PMCID: PMC3386067.
15. Wang, H., et al., One-step generation of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering. Cell 2013;153 (4):910-8.
16. Orlando SJ, Santiago Y, DeKelver RC, Freyvert Y, Boydston EA, Moehle EA, Choi VM, Gopalan SM, Lou JF, Li J, Miller JC, Holmes MC, Gregory PD, Urnov FD, Cost GJ. Zinc-finger nuclease-driven targeted integration into mammalian genomes using donors with limited chromosomal homology. Nucleic Acids Res. 2010 Aug;38(15):e152. doi: 10.1093/nar/gkq512. Epub 2010 Jun 8. PubMed PMID: 20530528; PubMed Central PMCID: PMC2926620.
17. Cristea S, Freyvert Y, Santiago Y, Holmes MC, Urnov FD, Gregory PD, Cost GJ. In vivo cleavage of transgene donors promotes nuclease-mediated targeted integration. Biotechnol Bioeng. 2013 Mar;110(3):871-80. doi: 10.1002/bit.24733.Epub 2012 Oct 23. PubMed PMID: 23042119.
18. Maresca M, Lin VG, Guo N, Yang Y. Obligate ligation-gated recombination (ObLiGaRe): custom-designed nuclease-mediated targeted integration through nonhomologous end joining. Genome Res. 2013 Mar;23(3):539-46. doi: 10.1101/gr.145441.112. Epub 2012 Nov 14. PubMed PMID: 23152450; PubMed Central PMCID: PMC3589542.
19. Guschin, D. Y., et al., A rapid and general assay for monitoring endogenous gene modification. Methods Mol Biol. 2010;649:247-56.
20. Huang, M. C., et al., A simple, high sensitivity mutation screening using Ampligase mediated T7 endonuclease I and Surveyor nuclease with microfluidic capillary electrophoresis. Electrophoresis. 2012 Mar;33(5):788-96
21. Yan W, Smith C, Cheng L. Expanded activity of dimer nucleases by combining ZFN and TALEN for genome editing. Sci Rep. 2013 Aug 7;3:2376. doi: 10.1038/srep02376.PubMed PMID: 23921522; PubMed Central PMCID: PMC3736171.
